Columnstore indexes provide a convenient and effective way to store large OLAP data natively in SQL Server. Despite being easy to implement, they are far from a "set it and forget it" feature!

Many considerations about how they are used can ensure that they outperform expectations while minimizing the compute needed to maintain them. This session takes us beyond the basics into their architecture, optimization, and the common mistakes made when implementing them.

How do write operations REALLY work in columnstore structures?
Why is data order imperative for performance?
Why is mixing OLTP and OLAP indexes not performing as expected?
Is it really cool to put columnstore indexes in memory?
What are all the system views that will tell me everything about columnstore indexes?

These questions (and more) will be answered! Strap in for a fast-paced, demo-heavy dive into the efficient storage of analytic data!